Drip and Ship Thrombolytic Therapy for Acute Ischemic Stroke
Use, Temporal Trends, and Outcomes
Background and Purpose—Interhospital transfer after use of intravenous tissue-type plasminogen activator (tPA) in acute stroke (drip and ship) is increasingly frequent. Small studies have suggested that drip and ship tPA is safe and increases rates of tPA use; however, little is known about real-world practice patterns. We sought to evaluate temporal trends in drip and ship tPA use and to compare the patient and hospital characteristics with that of conventional (front door) thrombolysis.
Methods—We analyzed data from 44 667 patients with ischemic stroke treated with intravenous tPA ≤3 hours of symptom onset in the Get With The Guidelines-Stroke program from April 2003 to October 2010 in 1440 hospitals. The main outcomes were the frequency of drip and ship tPA use over time, the characteristics of patients treated, and in-hospital outcomes, treatments, and complications.
Results—Among 44 667 patients treated with tPA, the drip and ship method was a common method (n=10 475; 23.5%), the use of which increased in parallel with the traditional tPA method over time. Patients treated by the drip and ship method differed significantly from front-door patients, with lower National Institutes of Health Stroke Scale scores when recorded (n=35 467). Crude in-hospital mortality (10.9%) and symptomatic intracranial hemorrhage (5.7%) in patients treated by the drip and ship method were slightly higher compared with those in front-door patients, and these differences persisted after risk adjustment.
Conclusions—Drip and ship tPA is common, used in 1 in 4 patients treated with tPA in the United States. Modest differences in mortality and symptomatic intracranial hemorrhage may be because of patient selection bias, post-tPA care differences, or unmeasured confounding. The drip and ship paradigm may facilitate widespread tPA use in patients with acute stroke.
Intravenous (IV) tissue-type plasminogen activator (tPA) is a treatment for acute ischemic stroke with proven benefits.1 Even though it is the only approved therapy for stroke, rates of tPA administration to all patients with ischemic stroke are exceedingly low.2 In many small- to medium-sized hospitals, the infrastructure required for the ongoing evaluation and monitoring of patients with post-tPA stroke may not be available.3 In addition, the absence of neurointensive or neurosurgical resources to manage post-tPA complications, such as symptomatic intracranial hemorrhage (sICH), may result in lower rates of tPA use.4
The drip and ship method, whereby tPA is administered locally in an emergency department followed by transfer to a stroke center, aims to increase the rates of tPA use and its timely administration. Because time to treatment is a critical determinant of efficacy,5 the drip and ship method is designed to facilitate earlier use of tPA, often in settings where tPA use would not otherwise be possible. Patients are frequently then transferred to stroke centers of higher capability for further evaluation, intervention, and monitoring. To date, there are limited data on the frequency of drip and ship administration of tPA for ischemic stroke in real-world practice.
Although there is increasing adoption of the use of the drip and ship paradigm in stroke systems,6–11 there are limited data on either the safety of this method or the nature of post-tPA care delivery in community-based practice. Prompt recognition and management of neurological and hemodynamic changes in the early hours after ischemic stroke are essential for safety, including during the medical transport period for patients treated by the drip and ship method.6 Available data using clinically abstracted variables on outcomes for drip and ship stroke have been limited to single-center and regional studies that may be subject to sampling bias and characteristics specific to individual regional networks.7–9 National data have been published, but they are derived from claims data lacking clinical detail.11 The generalizability of safety data, in particular, must be confirmed in patients treated in routine clinical practice because the drip and ship method becomes more commonly adopted.12 To address the need for a contemporary analysis of a large, clinical practice–derived data set, the US national Get With The Guidelines (GWTG)-Stroke registry was analyzed to determine (1) the frequency and temporal trends in the use of the drip and ship strategy, (2) differences in guideline-based care post-IV tPA, and (3) the association of the drip and ship method of tPA administration with outcomes in patients with acute ischemic stroke.
The GWTG-Stroke program is a US registry developed and maintained by the American Heart Association/American Stroke Association with the purpose of supporting quality and performance improvement, as well as serving as a platform for understanding the delivery of care and outcomes for patients with stroke and transient ischemic attack.13,14 The design, case ascertainment, and data collection methodologies have been reported previously.14 GWTG-Stroke uses a web-based interface patient management tool (Quintiles Real-World & Late Phase Research, Cambridge, MA) to collect clinical data and outcomes on consecutive patients admitted with stroke or transient ischemic attack, identified prospectively by clinical diagnosis or retrospectively using the International Classification of Diseases, Ninth Revision discharge codes, and confirmed by chart review. The GWTG-Stroke program was made available to any US hospital from 2003.13 The Duke Clinical Research Institute serves as the statistical coordinating center for the GWTG-Stroke Program. The Institutional Review Board at Duke University Health System approved the study.
For this analysis, data were included from April 1, 2003, to September 30, 2010, from fully participating hospitals in the GWTG-Stroke program during this time period. A total of 1 348 585 patients from 1582 hospitals were included in this initial sample. Major exclusion criteria were applied as follows: (1) sites with <30 patients (140 sites; 1773 patients) and (2) in-patient strokes (16 897). We further restricted the study population to patients with ischemic stroke who did not receive experimental therapies and who were not transferred out from the GWTG registry hospital to another acute care facility. Thus, our remaining study population included a total of 775 913 patients from 1440 hospitals. From this sample, 34 192 front-door patients (1216 sites) did not have documented reasons for no tPA use, received tPA within 3 hours of last known well time, and remained at the hospital where thrombolysis was administered versus 10 475 patients treated by the drip and ship method (640 sites) who received tPA at another hospital before transfer to the receiving GWTG hospital.
Patient-level data, including age, sex, race, last known well time, arrival time (defined as the time of arrival at the initial hospital for front-door patients and the transfer hospital for patients treated by the drip and ship method), medical history, diagnostic studies, tPA treatment initiation time, tPA complications, other in-hospital guideline–based treatments, discharge disposition, and destination, were all abstracted by trained hospital personnel. Patient location at the time of tPA administration (ie, front-door patients versus patients treated by the drip and ship method) was recorded. For patients treated by the drip and ship method, the initial National Institutes of Health Stroke Scale (NIHSS) score was measured on admission to the receiving hospital; in rare cases, it may have been performed before transfer if the GWTG staff were directly involved in the evaluation and tPA treatment decision, either in-person or remotely via telemedicine. The location of the patient at the time of the initial NIHSS score is not recorded in GWTG. Measures of guideline-based care, including antithrombotic and anticoagulation therapy, deep vein thrombosis prophylaxis, smoking cessation, statin administration, dysphagia screening, stroke education, and rehabilitation assessment, were also abstracted.
Hospital-level characteristics are also collected as part of the GWTG-Stroke program. These variables include bed size, geographic region, academic status, ischemic stroke volume, and primary stroke center certification by the Joint Commission.
Outcomes of Interest
We sought to describe trends in the type of tPA administration across the United States, by examining the overall frequency of drip and ship administration, as well as trends of its use over time. The primary patient–level outcomes of interest were in-hospital mortality, sICH, systemic hemorrhage, discharge to home for survivors, independent ambulation at discharge, and length of stay. Secondary outcomes of interest included a comparison by the tPA treatment strategy of guideline-based quality of care measures.
The frequency of drip and ship administration was determined by comparison of the total number of patients with ischemic stroke with that of patients with ischemic stroke treated with IV tPA. For patients with ischemic stroke, the denominator excluded those patients whose discharge information was missing or those who received experimental treatment. For patients treated with tPA, the denominator included those patients treated within a 3-hour window.
Univariate analyses were performed to compare front-door patients with patients treated by the drip and ship method across the variables of interest. Categorical variables are presented as either percentages or frequencies, whereas continuous variables are reported as medians with 25th and 75th percentiles. Pearson Χ2 tests were used for categorical variables, and Wilcoxon signed-rank sum tests were used for continuous variables. Differences between outcomes were assessed with multivariable logistic regression models using generalized estimating equations to account for in-hospital clustering of patients. The models were adjusted for prespecified patient characteristics, including age (0.0% missing), race (0.3% missing), sex (0.1% missing), arrival time off hours (5 pm to 7 am from Monday to Friday, Saturdays, and Sundays—0.0% missing), and medical history (atrial fibrillation, prosthetic heart valve, previous stroke/transient ischemic attack, coronary heart disease or previous myocardial infarction, carotid stenosis, peripheral vascular disease, hypertension, dyslipidemia, diabetes mellitus, and current smoking—9.3% missing), as well as hospital characteristics: region (0.0% missing), teaching status (4.4% missing), number of beds (6.0% missing), average annual volume of ischemic stroke (0.0% missing), average annual volume of drip and ship cases (0.0% missing), Joint Commission primary stroke center certification (0.0% missing), and rural location (0.0% missing). Missing values for patient characteristics were imputed to the most frequent category for categorical variables, with the exception of medical history. All medical history variables were imputed to none. Missing values for continuous variables were imputed to the median. If missing, hospital characteristics were not imputed, and thus, these missing values were excluded from the models.
Stroke severity is a critical prognostic factor in in-hospital mortality and short term outcomes,15,16 but the NIHSS score was missing in ≈30% of drip and ship cases and ≈15% of front-door cases. Therefore, we performed an additional sensitivity analysis evaluating risk-adjusted outcomes (covariates specified above, in addition to the NIHSS score) among patients in whom the NIHSS score was recorded (36 547).
Patients undergoing intra-arterial therapy generally have large vessel occlusions, and thus, they may have additional exposures associated with endovascular reperfusion therapies, general anesthesia, and other interventions to which the rest of the entire cohort was not exposed. Thus, we also ran the same risk-adjusted model for the overall tPA cohort, excluding those patients who received intra-arterial therapy.
Odds ratios and 95% confidence intervals were computed for all unadjusted and adjusted analyses. All statistical analyses were performed using SAS version 9.1 (SAS Institute, Inc, Cary, NC) software.
The study population consisted of 44 667 patients who were treated with IV tPA from 2003 to 2010. Their median age was 72 years, 49% were women, 12% were black, and 5% were Hispanic. In this study population, 23.5% of the patients were administered IV tPA via drip and ship approach. The percentage of patients presenting to GWTG hospitals and receiving tPA via drip and ship as a proportion of all patients with ischemic stroke eligible for tPA increased during the study period (4.0%–7.6%; P<0.0001; Figure [A]). Patients eligible for tPA were defined as those who did not have contraindications to tPA and were not under an investigational or experimental protocol for thrombolysis. However, the percentage of drip and ship tPA as a proportion of all patients treated with tPA consistently remained between 20% and 26% over all study years because overall rates of tPA use also increased in parallel (Figure [B]).
Patient-level characteristics of front-door patients versus patients treated by the drip and ship method are shown in Table 1. Compared with front-door patients, patients treated by the drip and ship method were younger, more often men, and more often white. Patients treated by the drip and ship method were more likely to arrive off hours (outside of Monday–Friday, 7 am–5 pm). Not surprisingly, the time from symptom onset to GWTG receiving hospital arrival was ≈5× longer for patients treated by the drip and ship method. Front-door patients more commonly had a history of atrial fibrillation/flutter, hypertension, and previous stroke or transient ischemic attack. NIHSS scores were missing more often in patients treated by the drip and ship method (29.7% versus 14.7%; P<0.0001), but when recorded, they were lower (11 versus 12; P<0.0001). Of note, the percentage of NIHSS score 0 to 4 was higher in patients treated by the drip and ship method compared with that in front-door patients (14.7% versus 9.1%; P<0.0001).
Patients treated by the drip and ship method were more likely to be admitted to sites with larger bed size, academic medical centers, higher volumes of annual ischemic stroke cases (273 versus 206; P<0.0001), and Joint Commission certified primary stroke centers than front-door patients. As a proportion, patients treated by the drip and ship method were more common in the Midwest (Table 1).
Clinical outcomes for front-door patients versus patients treated by the drip and ship method are presented in Table 2. The overall in-patient mortality rate in this cohort was 10%. sICH was defined as intracranial hemorrhage observed using computed tomography of the brain within 36 hours of tPA administration and clinical deterioration because of hemorrhage indicated in physician’s notes. It occurred in 5.4% of all patients administered with tPA. Life-threatening, serious systemic hemorrhage was defined as bleeding within 36 hours of IV tPA administration and >3 transfused units of blood within 7 days or discharge (whichever was earlier), and a physician note attributing bleeding problem as reason for transfusion occurred in 1.3%. Approximately 40% of patients were discharged home, and 38.7% of patients had achieved independent ambulation by discharge. Although statistically significant, the absolute differences in outcomes between the 2 groups were small (range, −0.4% to 1.26%).
In unadjusted analysis (Table 3), there was modest increased risk for sICH and mortality in patients treated by the drip and ship method and decreased likelihood of independent ambulation and discharge home. After adjustment for patient and hospital factors, including the NIHSS score (despite the recognized limitations of when NIHSS was acquired in patients treated by the drip and ship method), there remained modest differences in outcomes (Table 4) favoring front-door patients.
Because patients who undergo intra-arterial therapy are likely to have more severe stroke and unique exposures compared with patients receiving IV tPA only, such as anesthesia and endovascular therapy, we performed additional analyses to evaluate if these patients might be affecting the results. There were 1200 (3.6%) front-door patients and 707 (7.2%) patients treated by the drip and ship method who received endovascular therapy. Compared with patients who received IV tPA only, these patients had higher median NIHSS scores (17 versus 12; P<0.0001), higher rates of mortality (20% versus 10%; P<0.0001), and symptomatic hemorrhage (11% versus 5%; P<0.0001). After excluding the 1907 patients who underwent endovascular therapy from the adjusted models, including the NIHSS, there was little change on the rates of poor outcomes in patients treated by the drip and ship method (Tables I and II in the online-only Data Supplement).
Table 5 demonstrates differences in measures of guideline-based care between front-door patients and drip patients treated by the drip and ship method. Most quality measures (antithrombotic therapy, anticoagulation for atrial fibrillation/atrial flutter, deep vein thrombosis prophylaxis, and lipid lowering therapy) had higher adherence rates in patients treated by the drip and ship method; however, reflecting the large size of the database, the magnitudes of the differences were small (ranging from 1.1–3.3 percentage points). Dysphagia screening was less often adhered to in patients treated by the drip and ship method than in front-door patients (89.5% versus 91.6%; P<0.0001).
In this study of >40 000 patients with acute ischemic stroke treated with IV thrombolysis throughout the United States, drip and ship thrombolysis was commonly used and safe. From 2003 to 2010, overall tPA use has increased for patients with ischemic stroke, and the drip and ship approach has been an important contributor to this increase, accounting for 1/4th of all patients who received acute thrombolysis. Our results demonstrate that the proportion of all patients with ischemic stroke who received drip and ship tPA increased from 2003 to 2010 and that this increase paralleled the increase in the proportion of front-door eligible patients who received tPA. These observations suggest that among hospitals participating in the GWTG Stroke registry, the drip and ship approach continued to be an important means of delivering acute stroke therapy from stroke systems of care perspective. Further analysis is needed to identify and understand barriers to the drip and ship approach in those geographic regions where this method of tPA administration is less common.
The lack of an onsite acute stroke team or neurointensive care is the potential reason why an initial receiving facility may rely on drip and ship tPA administration. During the study period, patients treated by the drip and ship method, as a percentage of patients eligible for tPA, nearly doubled. Our data help address lingering concerns that patients treated by the drip and ship method are at substantially increased risk of harm or may receive less effective post-tPA care, by demonstrating that unadjusted absolute differences in mortality (1.23%) and symptomatic hemorrhage (0.5%) are small. Relative to the likelihood of potential improvement from tPA administration, the risk of poor outcome is low, supporting the use of the drip and ship strategy.
Furthermore, several factors suggest that any absolute increase in poor outcome seen in this analysis may be an overestimate. The baseline NIHSS score, as a measure of baseline severity essential for adequate adjustment,15,16 may be underestimated in patients treated by the drip and ship method in this analysis, biasing outcomes to seem worse for patients treated by the drip and ship method. In GWTG, the NIHSS score is almost always recorded at the receiving hospital, long after tPA administration was started at the referring hospital. With a 4.8-fold difference in median last known well to the arrival at the GWTG receiving hospital (249 versus 51 minutes) for patients treated by the drip and ship method, it is likely that clinical improvement occurred in a substantial number of patients treated by the drip and ship method after early IV tPA administration at the sending hospital. Early improvement after tPA was well documented in the original National Institute of Neurological Disorders and Stroke (NINDS) tPA trial.17 For patients treated by the drip and ship method, the captured NIHSS score is usually at the receiving hospital, post-tPA infusion, likely underestimating true baseline severity. This bias does not exist for front-door patients. Further supporting this observation is the increased number of NIHSS score from 0 to 4 in patients in the drip and ship cohort compared with front-door patients, although the evaluation of this cohort was not prespecified. This analysis highlights the need to document and report baseline NIHSS scores both at sending and receiving hospitals for patients treated by the drip and ship method with stroke to perform appropriate baseline risk adjustment.
Confounding by indication may also occur in patients treated by the drip and ship method with regard to likelihood of large vessel occlusion and increased stroke severity. The proportion of patients treated by the drip and ship method who underwent intra-arterial therapy was twice the number of that of patients who received intra-arterial therapy in the front-door population, suggesting a higher frequency of large vessel occlusions in this group that increases the risk of a poor outcome compared with patients with a similar NIHSS score but no proximal occlusion. Finally, patients treated by the drip and ship method were not eliminated from the analysis for IV tPA >3 hours. Despite these potential concerns, the possibility remains that patients treated by the drip and ship method may have worse outcomes, and further study, with adequate pre-tPA baseline risk adjustment, is needed.
This analysis used front-door administration of tPA as the comparison arm. An alternative to drip and ship tPA is either a long delay to treatment or no thrombolysis at all. Given the compelling data on the benefits of early thrombolysis demonstrated in clinical trials and in the GWTG Stroke registry,18 it is likely that the small observed increased risk of sICH or mortality in this drip and ship population is offset by earlier tPA initiation. To fully address this question, future studies must either randomize patients to drip and ship versus usual care (which seems unlikely) or collect additional baseline data to better characterize these patients and capture other currently unmeasured potential confounders. These data should include an assessment of NIHSS scores at both shipping and receiving hospitals and allow linkages of patients between hospitals, both of which were not possible in our study. There is limited information about the incidence of neurological deterioration during transfer and adherence to post-tPA guidelines during emergency transport, and further investigation in this area is urgently needed.
One previous national US study has examined drip and ship tPA administration,11 finding that the drip and ship method was used in 17% of tPA-treated cases and that in-hospital mortality outcomes in patients treated by the drip and ship method were not different from those in front-door patients after adjusting for age, sex, risk factors, and hospital teaching status. However, this analysis was assembled using billing diagnosis codes without clinical validation and did not include any adjustment for baseline stroke severity. In contrast, our analysis leverages a widely used contemporary quality improvement stroke database in which baseline clinical characteristics are directly entered by individual hospitals and in which previous studies have demonstrated a representative sample compared with Medicare fee-for-service ischemic stroke patients.19 However, in both our analysis and the previous study, drip and ship use, as well as tPA use in general, increased over time. Further research is required to understand factors and policies that facilitate the creation and maintenance of drip and ship programs, as well as regional and socioeconomic disparities that may exist with regard to access to drip and ship tPA administration.
Our findings extend previous observations about the relationship between the method of tPA delivery and clinical outcomes from previous single-center and regional studies,7–11 but in a larger, nationwide study. Recent analyses demonstrated that clinical outcomes were the same or better in patients treated with drip and ship than those in front-door patients.11,20 Previous smaller studies may not have had sufficient statistical power to detect small differences. Reassuringly, there were no consistent patterns of difference in postacute guideline–based care (Table 5). Further study, with adjustment for initial pre-tPA administration stroke severity, is imperative to understand whether there is any absolute increase in poor outcome among patients treated by the drip and ship method. Telemedicine may be used with considerable variability in this cohort, and the role of telemedicine in facilitating safe and effective transfer of patients needs to be clarified.20 Finally, the role of ancillary aspects of drip and ship care that may make tPA administration even safer, including emergency transport and neurocritical care, is an important area of future study.
There are several potential limitations that are important to note. Hospitals participating in the GWTG-Stroke program are self-selected and likely motivated to improve stroke care quality. Therefore, the findings may not be generalizable to other US hospitals; however, comparisons of GWTG-Stroke registry with Medicare registries indicate that patients in the GWTG registry are representative of the overall stroke population in the United States.19 The quality of the data is dependent on the accuracy of data abstraction at each center. The GWTG-Stroke program optimizes training of abstractors, standardizes definitions and coding instructions, and uses a series of internal quality checks and data audits to verify accuracy.14 The results of previous audits have demonstrated high levels of quality.21 We did not evaluate the safety and efficacy of IV tPA in the 3- to 4.5-hour time window because not all US hospitals offer tPA in that extended window which is guideline recommended but not Food and Drug Administration approved and because our study start precedes the recommendation for extended window treatment by 6 years. We also did not have data about long-term functional outcomes. Future studies will be needed to determine the adoption of the drip and ship paradigm and its safety and efficacy when it is used for patients treated beyond 3 hours.
In the largest study of its kind to date, using a national registry capturing current clinical practice in acute stroke, the drip and ship method of IV tPA treatment accounts for 1 in 4 of the patients with acute ischemic stroke treated with tPA. These results suggest that the drip and ship method is an important, frequently used strategy for tPA administration. Outcomes in patients treated by the drip and ship method showed modest differences compared with those of directly admitted patients treated with tPA, but these results may reflect patient selection bias, post-tPA care differences, or unmeasured confounding. Further baseline data capture and additional targeted quality improvement programs and performance measures may help improve the outcomes in patients treated via the drip and ship approach.
Sources of Funding
The Get With The Guidelines (GWTG)-Stroke Program is provided by the American Heart Association/American Stroke Association (AHA/ASA). The GWTG-Stroke program is currently supported in part by a charitable contribution from Janssen Pharmaceutical Companies of Johnson & Johnson. The GWTG-Stroke Program had been funded in the past through support from Boehringer-Ingelheim, Merck, Bristol-Myers Squib/Sanofi Pharmaceutical Partnership, and the AHA Pharmaceutical Roundtable. The industry sponsors of the GWTG-Stroke program had no role in the design and conduct of the study; in the collection, analysis, and interpretation of the data; or in the preparation, review, or approval of the article, and decision to submit the article for publication.
Dr Sheth is a member of the Get with the Guidelines (GWTG)-Stroke Clinical Workgroup, and he is a Co-Principal Investigator and Executive Committee member for Glyburide Advantage in Malignant Edema and Stroke-Remedy Pharmaceuticals (GAMES-RP), a phase II–trial to prevent swelling in patients with large stroke, funded by Remedy Pharmaceuticals, Inc. Dr Smith is a member of the GWTG-Stroke Workgroup. Dr Kleindorfer discloses speaking engagements. Dr Fonarow is a member of the GWTG Steering Committee; receipt of research support (to the institution) from Patient-Centered Outcomes Research Institute, and he is an employee of the University of California that holds a patent on retriever devices for stroke. Dr Schwamm is the chair of the GWTG-Stroke Clinical Workgroup, a principal investigator of the National Institutes of Health–funded MR WITNESS (A Study of Intravenous Thrombolysis With Alteplase in MRI-Selected Patients) trial of extended window thrombolysis for which Genentech provides supplemental site payments and alteplase free of charge, a member of the international steering committee of the Desmoteplase in Acute Ischemic Stroke (DIAS) 3 and 4 trials of extended window thrombolysis, and the director of Massachusetts General Hospital (MGH) TeleHealth. The MGH provides a broad array of telehealth services to hospitals in New England, including telestroke-enabled thrombolysis. Dr Grau-Sepulveda reports no conflicts.
Guest Editor for this article was James Grotta, MD.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.114.007506/-/DC1.
- Received September 19, 2014.
- Revision received November 17, 2014.
- Accepted December 2, 2014.
- © 2015 American Heart Association, Inc.
- Jauch EC,
- Saver JL,
- Adams HP Jr,
- Bruno A,
- Connors JJ
- 2.↵Paul Coverdell Prototype Registries Writing Group. Acute stroke care in the US: results from 4 pilot prototypes of the paul coverdell national acute stroke registry. Stroke. 2005;36:1232–1240. doi: 10.1161/01.STR.0000165902.18021.5b.
- Higashida R,
- Alberts MJ,
- Alexander DN,
- Crocco TJ,
- Demaerschalk BM,
- Derdeyn CP,
- et al
- Asaithambi G,
- Chaudhry SA,
- Hassan AE,
- Rodriguez GJ,
- Suri MF,
- Qureshi AI
- Tekle WG,
- Chaudhry SA,
- Hassan AE,
- Rodriguez GJ,
- Suri MF,
- Qureshi AI
- Alberts MJ,
- Wechsler LR,
- Jensen ME,
- Latchaw RE,
- Crocco TJ,
- George MG,
- et al
- Schwamm LH,
- Fonarow GC,
- Reeves MJ,
- Pan W,
- Frankel MR,
- Smith EE,
- et al
- Fonarow GC,
- Saver JL,
- Smith EE,
- Broderick JP,
- Kleindorfer DO,
- Sacco RL,
- et al
- Hemmen TM,
- Ernstrom K,
- Raman R
- Fonarow GC,
- Zhao X,
- Smith EE,
- Saver JL,
- Reeves MJ,
- Bhatt DL,
- et al
- Reeves MJ,
- Fonarow GC,
- Smith EE,
- Pan W,
- Olson D,
- Hernandez AF,
- et al
- Pervez MA,
- Silva G,
- Masrur S,
- Betensky RA,
- Furie KL,
- Hidalgo R,
- et al